Skip to content
View atharshlakshmi's full-sized avatar

Block or report atharshlakshmi

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
atharshlakshmi/README.md

Hi, I'm Lakshmi! ⭐️

Core Ambition: Building a career dedicated to contributing directly to the field of AI Safety and Responsible AI.

I am a Data Science & AI major with a passion for building trustworthy systems. As I explore my interests, I'm developing projects focused on alignment, accessibility, and robust machine learning.

🛠️ Technical Stack

Category Tools & Frameworks
AI / Machine Learning PyTorch Scikit-Learn Pandas NumPy Streamlit
Backend & Data Python FastAPI SQL

🔍 Research & Projects

🛡️ AI Safety & Responsible AI

  • SandBagDetect (2026) – Research on consistency probing for detecting sandbagging in LLMs.
  • Clean Talk (2025) – A Guardrail API for enforcing Responsible AI policies.

📖 Research

  • GOALIE (2025) – Exploring LLMs in Embodied Environments for goal decomposition.

🏆 Hackathon Highlights

  • AURA (2026) – Finalist | Intent-driven browser designed for accessibility.
  • Common Ground (2025) – 2nd Place | Visual novel for Deaf etiquette education.
  • NOVA (2026) – Node-oriented visualization for AI in education.
  • validO (2025) – ML system for evaluating Google Local review quality.

📬 Connect with Me

LinkedIn

Pinned Loading

  1. clean-talk clean-talk Public

    Jupyter Notebook

  2. SandbagDetect SandbagDetect Public

    Jupyter Notebook 1

  3. validO validO Public

    Jupyter Notebook

  4. klk645445/SignLanguageVisualNovel klk645445/SignLanguageVisualNovel Public

    TypeScript 2